Lock down kwargs in offsets signatures #17458

jbrockmendel · 2017-09-07T04:04:38Z

Explicitly specify allowed kwargs in __init__ for Week, WeekOfMonth, LastWeekOfMonth and QuarterOffset.

Ref #17176

Intended to be orthogonal/complementary to #17450.

Will add tests after confirming we're on the same page.

closes #xxxx
tests added / passed
passes git diff upstream/master -u -- "*.py" | flake8 --diff
whatsnew entry

codecov · 2017-09-07T07:59:53Z

Codecov Report

Merging #17458 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #17458      +/-   ##
==========================================
- Coverage   91.15%   91.14%   -0.02%     
==========================================
  Files         163      163              
  Lines       49591    49588       -3     
==========================================
- Hits        45207    45195      -12     
- Misses       4384     4393       +9

Flag	Coverage Δ
#multiple	`88.92% <100%> (-0.01%)`	⬇️
#single	`40.25% <26.66%> (-0.07%)`	⬇️

Impacted Files	Coverage Δ
pandas/tseries/offsets.py	`97.14% <100%> (-0.01%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/frame.py	`97.72% <0%> (-0.1%)`	⬇️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 20fee85...5e276c9. Read the comment docs.

codecov · 2017-09-07T08:00:04Z

Codecov Report

Merging #17458 into master will decrease coverage by 0.02%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #17458      +/-   ##
==========================================
- Coverage   91.26%   91.24%   -0.03%     
==========================================
  Files         163      163              
  Lines       49872    49978     +106     
==========================================
+ Hits        45514    45600      +86     
- Misses       4358     4378      +20

Flag	Coverage Δ
#multiple	`89.04% <100%> (ø)`	⬆️
#single	`40.24% <30.35%> (-0.06%)`	⬇️

Impacted Files	Coverage Δ
pandas/tseries/offsets.py	`97.16% <100%> (+0.01%)`	⬆️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/indexing.py	`93% <0%> (-0.98%)`	⬇️
pandas/core/series.py	`94.89% <0%> (-0.16%)`	⬇️
pandas/core/frame.py	`97.74% <0%> (-0.09%)`	⬇️
pandas/core/generic.py	`92.03% <0%> (-0.09%)`	⬇️
pandas/core/common.py	`91.42% <0%> (ø)`	⬆️
pandas/core/ops.py	`91.76% <0%> (ø)`	⬆️
pandas/core/internals.py	`94.38% <0%> (ø)`	⬆️
pandas/core/indexes/period.py	`92.78% <0%> (+0.01%)`	⬆️
... and 8 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 81694dc...086b485. Read the comment docs.

jreback

can you add a whatsnew note in apbreaking (maybe a sub-section), where you note which thinks are changing, and how it could potentitally affect the user. In future PR's can expand this as things are added.

…fset_sigs

jreback · 2017-09-10T14:47:02Z

doc/source/whatsnew/v0.21.0.txt

+Restricting DateOffset Keywords
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Previously, ``DateOffset`` subclasses allowed arbitrary keyword arguments which could lead to unexpected behavior.  Now, subclasses :class:`Week`, :class:`WeekOfMonth`, :class:`LastWeekOfMonth`, :class:`QuarterOffset`, :class:`QuarterEnd`, :class:`QuarterBegin`, :class:`BQuarterEnd`, and :class:`BQuarterBegin`, will allow only specific arguments.  (:issue:`17176`).


aren't Month/MonthEnd included here? why not all classes?

aren't Month/MonthEnd included here?

Not as of now, no.

why not all classes?

Wanted to change them a couple at a time, err on the side of doing too little. Now that #17486 is resolved I can push forward on a few more. You want to just get them all in one go?

yes for this type of change am ok with doing it in one fell swoop. On the whatsnew just highlite the change generally (e.g. give a single example), then list the classes that are affected (or if its all, then just say all). imagine a user who just 'uses' pandas. They want to quickly get to: does this change affect me, if so, what do I need to do to fix.

jreback

in-line comment

jreback · 2017-09-23T17:37:04Z

ok, rebase this and i'll look again.

jbrockmendel · 2017-09-23T19:23:59Z

Will rebase, but I haven't gotten around to addressing your earlier comments, so this one isn't ready yet. #17564, #17585, #17363, #17543, and #17342 should be good to go.

…fset_sigs Lock down kwargs in Week.__init__

pep8speaks · 2017-09-24T03:31:13Z

Hello @jbrockmendel! Thanks for updating the PR.

Cheers ! There are no PEP8 issues in this Pull Request. 🍻

Comment last updated on October 06, 2017 at 03:52 Hours UTC

jbrockmendel · 2017-09-24T03:32:49Z

OK, as requested, the new commits cover all of the existing classes.

TomAugspurger

lgtm. Feel free to merge @jreback

jreback · 2017-09-25T23:25:13Z

doc/source/whatsnew/v0.21.0.txt

@@ -387,6 +387,13 @@ New Behavior

 Additionally, DataFrames with datetime columns that were parsed by :func:`read_sql_table` and :func:`read_sql_query` will also be localized to UTC only if the original SQL columns were timezone aware datetime columns.

+.. _whatsnew_0210.api.restricting_dateoffset_keywords
+
+Restricting DateOffset Keywords


ok, since we made this simpler, move to just a bullet point in Other API changes section.

jreback · 2017-09-25T23:27:05Z

pandas/tseries/offsets.py

@@ -1121,7 +1125,7 @@ class SemiMonthOffset(DateOffset):
    _default_day_of_month = 15
    _min_day_of_month = 2

-    def __init__(self, n=1, day_of_month=None, normalize=False, **kwds):
+    def __init__(self, n=1, normalize=False, day_of_month=None):


is this consistent with other freq signatures?

All the other signatures are (self, n=1, normalize=False, .... In the status quo this is an outlier.

yes that's fine. see my comment below though.

this is fine, in an ideal world we would change it but can't now

jreback · 2017-09-25T23:28:03Z

pandas/tseries/offsets.py

    """

-    def __init__(self, n=1, normalize=False, **kwds):
+    def __init__(self, n=1, normalize=False, weekday=None):


is there a test for this as weekday=None? isn't that an error?

It is an error, though a different one in Py2 vs Py3. In the status quo not passing a weekday explicitly will raise a KeyError. Default to Monday I guess?

no if weekday is None, just raise a value error, in reality this should actually be the first (and a required parameter), but I guess that would be hard to change. Don't set defaults, raise

Are you suggesting it should be __init__(self, weekday, n=1, normalize=False)? I see some appeal to that, but there is a lot of symmetry with other signatures all starting with (self, n=1, normalize=False, ...).

see my comment above. this is ok for now. actually weekday is optional, so this is fine.

Two things. First, here and in other places you say "this is fine" but I don't know whether you're referring to the PR or the thing the PR is fixing. Based on the go-ahead you've given below to look over timeseries.rst I'm assuming you mean the PR version is fine, just heads up for ways I'm liable to get confused next time around.

actually weekday is optional,

Under the status quo, failing to pass weekday raises a KeyError. In the PR, not passing anything will raise a ValueError in py2 and TypeError in py3. (The same errors would result in the status quo if a user passed None explicitly)

I never mean this entire PR is fine. If that was the case I wouldn't have any other comments. Individual comments are revering to the specific line/section where I make it.

jreback · 2017-09-25T23:28:32Z

pandas/tseries/offsets.py

@@ -1829,13 +1835,14 @@ class QuarterOffset(DateOffset):
    # TODO: Consider combining QuarterOffset and YearOffset __init__ at some
    #       point

-    def __init__(self, n=1, normalize=False, **kwds):
+    def __init__(self, n=1, normalize=False, startingMonth=None):


can you make an issue that this should be deprecated to starting_month (and prob others like this)

jreback

looks fine, can you pass thru the timeseries.rst docs and if anything needs changing there (should be ok), but who knows

…fset_sigs

jreback · 2017-09-30T21:50:54Z

pandas/tests/tseries/data/dateoffset_0_19_2.pickle

+S'YearBegin'
+p1
+ccopy_reg
+_reconstructor


not what i asked for at all

pls use the generate_legacy_data script
we have removed all of these ad hoc pickle things
we r not adding them back

You asked me to run a script that doesn't work, then to troubleshoot it until it does. I found existing tests and followed their example. Please cut me some darn slack.

we have removed all of these ad hoc pickle things

test_pickle_v0_15_2 and test_pickle_compat_0_14_1 might lead a naive reader (such as myself) to suspect otherwise.

you can use pandas versions to do that ONLY in that script to do that.

This sentence confuses me.

If you make changes [to the generation script] then they should be checked in as part of this PR.

I have no interest in making changes to the non-functional generation script.

I have no interest in making changes to the non-functional generation script.

well it works fine according to the instructions I gave and has been used for quite a while.

test_pickle_v0_15_2 and test_pickle_compat_0_14_1 might lead a naive reader

well if you look at what exists, the vast majority are in an organized format. The remaining ones should actually be converted to this format.

(pandas) bash-3.2$ find pandas -name '*pickle*' pandas/compat/__pycache__/pickle_compat.cpython-36.pyc pandas/compat/pickle_compat.py pandas/io/__pycache__/pickle.cpython-36.pyc pandas/io/pickle.py pandas/tests/indexes/data/mindex_073.pickle pandas/tests/indexes/data/multiindex_v1.pickle pandas/tests/io/__pycache__/test_pickle.cpython-36-PYTEST.pyc pandas/tests/io/data/categorical_0_14_1.pickle pandas/tests/io/data/categorical_0_15_2.pickle pandas/tests/io/data/legacy_pickle pandas/tests/io/data/legacy_pickle/0.10.1/AMD64_windows_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.10.1/x86_64_linux_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.11.0/0.11.0_x86_64_linux_3.3.0.pickle pandas/tests/io/data/legacy_pickle/0.11.0/x86_64_linux_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.11.0/x86_64_linux_3.3.0.pickle pandas/tests/io/data/legacy_pickle/0.12.0/0.12.0_AMD64_windows_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.12.0/0.12.0_x86_64_linux_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_AMD64_windows_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_i686_linux_2.6.5.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_i686_linux_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_i686_linux_3.2.3.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_x86_64_darwin_2.7.5.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_x86_64_darwin_2.7.6.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_x86_64_linux_2.7.3.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_x86_64_linux_2.7.8.pickle pandas/tests/io/data/legacy_pickle/0.13.0/0.13.0_x86_64_linux_3.3.0.pickle pandas/tests/io/data/legacy_pickle/0.14.0/0.14.0_x86_64_darwin_2.7.6.pickle pandas/tests/io/data/legacy_pickle/0.14.0/0.14.0_x86_64_linux_2.7.8.pickle pandas/tests/io/data/legacy_pickle/0.14.1/0.14.1_x86_64_darwin_2.7.12.pickle pandas/tests/io/data/legacy_pickle/0.14.1/0.14.1_x86_64_linux_2.7.8.pickle pandas/tests/io/data/legacy_pickle/0.15.0/0.15.0_x86_64_darwin_2.7.12.pickle pandas/tests/io/data/legacy_pickle/0.15.0/0.15.0_x86_64_linux_2.7.8.pickle pandas/tests/io/data/legacy_pickle/0.15.2/0.15.2_x86_64_darwin_2.7.9.pickle pandas/tests/io/data/legacy_pickle/0.16.0/0.16.0_x86_64_darwin_2.7.9.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_AMD64_windows_2.7.10.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_AMD64_windows_3.4.3.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_x86_64_darwin_2.7.10.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_x86_64_darwin_2.7.9.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_x86_64_darwin_3.4.3.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_x86_64_linux_2.7.10.pickle pandas/tests/io/data/legacy_pickle/0.16.2/0.16.2_x86_64_linux_3.4.3.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.0_AMD64_windows_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.0_AMD64_windows_3.4.4.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.0_x86_64_darwin_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.0_x86_64_darwin_3.4.4.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.0_x86_64_linux_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.0_x86_64_linux_3.4.4.pickle pandas/tests/io/data/legacy_pickle/0.17.0/0.17.1_AMD64_windows_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.17.1/0.17.1_AMD64_windows_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.17.1/0.17.1_x86_64_darwin_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.18.0/0.18.0_AMD64_windows_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.18.0/0.18.0_AMD64_windows_3.5.1.pickle pandas/tests/io/data/legacy_pickle/0.18.0/0.18.0_x86_64_darwin_2.7.11.pickle pandas/tests/io/data/legacy_pickle/0.18.0/0.18.0_x86_64_darwin_3.5.1.pickle pandas/tests/io/data/legacy_pickle/0.18.1/0.18.1_x86_64_darwin_2.7.12.pickle pandas/tests/io/data/legacy_pickle/0.18.1/0.18.1_x86_64_darwin_3.5.2.pickle pandas/tests/io/data/legacy_pickle/0.19.2/0.19.2_x86_64_darwin_2.7.12.pickle pandas/tests/io/data/legacy_pickle/0.19.2/0.19.2_x86_64_darwin_3.6.1.pickle pandas/tests/io/test_pickle.py pandas/tests/tseries/data/cday-0.14.1.pickle pandas/tests/tseries/data/dateoffset_0_15_2.pickle

As you know pandas is a very large library. If things do not have process & procedure around them, then future generations will have pain. We have worked pretty hard (and you are contributing to this), to reduce technical debt.

jreback · 2017-10-01T20:44:40Z

see #17733

jreback · 2017-10-01T20:52:40Z

@jbrockmendel
since we really are locking down signatures. I would write an old (but acceptable signature) and add to generate_legacy_storage_files (e.g. maybe with the kwargs or something). I think this will all work, just want to confirm.

jreback · 2017-10-01T20:53:47Z

I think I covered this case pandas/tests/tseries/data/dateoffset_0_15_2.pickle as well, so can remove that (now I didn't run it for 0.15.2 so you can certainly do that).

…fset_sigs

jbrockmendel · 2017-10-02T04:22:30Z

Thanks for fixing the script. Added a file for 0.20.3 for kicks.

So since test_pickle_v0_15_2 isn't an example for what you want a test to look like I'm guessing iterate over the files unpickling them and checking that the data['offsets'] matches the off dict defined in generate_legacy_storage_files.create_data?

I think I covered this case pandas/tests/tseries/data/dateoffset_0_15_2.pickle as well, so can remove that (now I didn't run it for 0.15.2 so you can certainly do that).

There's some subject-verb ambiguity here. What did you cover/run? What can be removed?

I would write an old (but acceptable signature) and add to generate_legacy_storage_files (e.g. maybe with the kwargs or something)

So go into generate_legacy_storage_files and add something to the off dict and generate a new pickle file? But wouldn't that invalidate the existing files?

jreback · 2017-10-03T20:13:06Z

So go into generate_legacy_storage_files and add something to the off dict and generate a new pickle file? But wouldn't that invalidate the existing files?

no you can just generate a new one. We only compare to things that exist in the pickle AND in the generate script. So as long as you simply 'add' things its ok (e.g. don't rename or delete).

I added an offset section, you can simply add moar things to it (or even another version of it that say initializes with keyword args or somesuch).

…fset_sigs

jbrockmendel · 2017-10-03T21:02:56Z

OK, so I generated a new file by running the script under 0.20.3. Then... what? The existing pickle tests in tests.tseries.test_offsets are precedent you've explicitly said not to follow.

jreback · 2017-10-04T10:36:37Z

pandas/tests/tseries/test_offsets.py

@@ -502,6 +502,31 @@ def test_pickle_v0_15_2(self):
        #
        tm.assert_dict_equal(offsets, read_pickle(pickle_path))

+    def test_pickle_v0_20_3(self):


none of this is needed

these go in the generate_legacy_storage_files script instead. (which you then use to generate). I think the instructions are now pretty clear. If they are not, feel free to update them as well (at the top of that file).

OK, just removed that test. As to the instructions, I'm still unclear on where the actual test goes and what it looks like.

the test is generic it runs ALL of the pickle files

that’s the point you don’t actually need to construct a test just put a new pickle file

OK. test_pickle_v0_20_3 has been removed, so it sounds like we're good to go.

jreback

did you make changes in generate_legacy_storage_files? or just generate 0.20.3? I wanted you to update that with passing keywords so we can cover that as well, e.g.

ideally we would cover kwargs for most of the offsets.

In [3]: offsets.SemiMonthBegin(1, normalize=True, day_of_month=4)
Out[3]: <SemiMonthBegin: day_of_month=4>

This then assures that we have backcompat.

jbrockmendel · 2017-10-06T03:53:59Z

Just added a handful of new examples to generate_legacy_storage_files, updated the 0.20.3 file, created a 0.19.2 file. Attempts to run under py3 raised

  File "pandas/pandas/tests/io/generate_legacy_storage_files.py", line 40, in <module>
    from pandas import (Series, DataFrame, Panel,
  File ".virtualenvs/pandas_0.19/lib/python3.6/site-packages/pandas/__init__.py", line 85, in <module>
    v = get_versions()
  File ".virtualenvs/pandas_0.19/lib/python3.6/site-packages/pandas/_version.py", line 23, in get_versions
    return json.loads(version_json)
AttributeError: module 'json' has no attribute 'loads'

jreback · 2017-10-06T10:33:44Z

also seems we should update the doc-strings for the offsets now that the signature are ok, xref #8435

jreback · 2017-10-06T10:36:03Z

File "pandas/pandas/tests/io/generate_legacy_storage_files.py", line 40, in
from pandas import (Series, DataFrame, Panel,
File ".virtualenvs/pandas_0.19/lib/python3.6/site-packages/pandas/init.py", line 85, in
v = get_versions()
File ".virtualenvs/pandas_0.19/lib/python3.6/site-packages/pandas/_version.py", line 23, in get_versions
return json.loads(version_json)
AttributeError: module 'json' has no attribute 'loads'

This is because we have a top-level file called json.py for backwards compat (was a c-module), which shadows the python json library, and the version things actually import this. So in this case you need to temporarily edit the pandas/_version.py file (in the version you are actually using) to avoid use of json (just change it to return a dict). a bit of a hack, but I have found no other way of fixing this (and it doesn't affect the released version, just running it like this).

jreback · 2017-10-06T10:38:02Z

ok this looks fine. Can you update the doc-strings? (also can do in another PR). now that the keywords are there they should be documented. lmk

jbrockmendel · 2017-10-06T15:29:53Z

Can you update the doc-strings? (also can do in another PR).

This PR has been exhausting. I'll put this on the TODO list.

jreback · 2017-10-06T15:37:57Z

@jbrockmendel hahah, ok thanks1

Lock down kwargs in offsets signatures

5e276c9

jreback requested changes Sep 7, 2017

View reviewed changes

jreback added the Frequency DateOffsets label Sep 7, 2017

jbrockmendel added 2 commits September 7, 2017 07:57

whatsnew note

0183e50

Merge branch 'master' of https://github.com/pandas-dev/pandas into of…

e33626e

…fset_sigs

jreback requested changes Sep 10, 2017

View reviewed changes

jreback mentioned this pull request Sep 14, 2017

Fix bug where offset.copy() != offset #17452

Merged

4 tasks

jreback requested changes Sep 17, 2017

View reviewed changes

jbrockmendel mentioned this pull request Sep 23, 2017

Bitesize offsets #17318

Merged

1 task

jbrockmendel added 7 commits September 23, 2017 19:12

Merge branch 'master' of https://github.com/pandas-dev/pandas into of…

1377df1

…fset_sigs Lock down kwargs in Week.__init__

restrict YearOffset to month kwarg

ed78a4a

Restrict SemiMonthOffset kwargs to day_of_month

36290d2

Lock down kwargs in FY5253 and FY5253Quarter

4443d6c

lock down kwargs in BusinessDay, CustomBusinessMonthEnd

8cb6b66

Lockdown kwargs in remaining DateOffset subclasses

fe7bb56

Briefer WhatsNew, add missing kwds attrs

3daab66

whitespace fixups

aefb68c

TomAugspurger approved these changes Sep 25, 2017

View reviewed changes

jreback requested changes Sep 25, 2017

View reviewed changes

jbrockmendel mentioned this pull request Sep 26, 2017

tslibs TODO list #17652

Closed

59 tasks

jbrockmendel added 2 commits September 25, 2017 17:42

Move doc section to other api changes bullet point

1e06d99

Merge branch 'master' into offset_sigs

7be31da

jreback approved these changes Sep 26, 2017

View reviewed changes

jreback added this to the 0.21.0 milestone Sep 26, 2017

Merge branch 'master' of https://github.com/pandas-dev/pandas into of…

e93de50

…fset_sigs

jreback requested changes Sep 30, 2017

View reviewed changes

Merge branch 'master' of https://github.com/pandas-dev/pandas into of…

7426265

…fset_sigs

jbrockmendel added 2 commits October 1, 2017 21:23

New version of test_pickle_v0_20_3

aee75de

flake8 whitespace fixup

cc5a71a

Merge branch 'master' of https://github.com/pandas-dev/pandas into of…

521a8d0

…fset_sigs

jreback requested changes Oct 4, 2017

View reviewed changes

remove test per reviewer instruction

217a558

jreback requested changes Oct 5, 2017

View reviewed changes

Add offsets with kwds to generate_legacy_storage_files

086b485

jreback approved these changes Oct 6, 2017

View reviewed changes

jreback added this to the 0.21.0 milestone Oct 6, 2017

jreback merged commit e63c935 into pandas-dev:master Oct 6, 2017

ghost pushed a commit to reef-technologies/pandas that referenced this pull request Oct 16, 2017

Lock down kwargs in offsets signatures (pandas-dev#17458)

77d9e49

jbrockmendel mentioned this pull request Oct 18, 2017

PERF: enable caching on expensive offsets #17914

Closed

jbrockmendel deleted the offset_sigs branch October 30, 2017 16:22

alanbato pushed a commit to alanbato/pandas that referenced this pull request Nov 10, 2017

Lock down kwargs in offsets signatures (pandas-dev#17458)

81eddbf

No-Stream pushed a commit to No-Stream/pandas that referenced this pull request Nov 28, 2017

Lock down kwargs in offsets signatures (pandas-dev#17458)

57c14e2

Lock down kwargs in offsets signatures #17458

Lock down kwargs in offsets signatures #17458

Conversation

jbrockmendel commented Sep 7, 2017

codecov bot commented Sep 7, 2017 • edited Loading

Codecov Report

codecov bot commented Sep 7, 2017 • edited Loading

Codecov Report

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback left a comment

Choose a reason for hiding this comment

jreback commented Sep 23, 2017

jbrockmendel commented Sep 23, 2017

pep8speaks commented Sep 24, 2017 • edited Loading

Comment last updated on October 06, 2017 at 03:52 Hours UTC

jbrockmendel commented Sep 24, 2017

TomAugspurger left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback commented Oct 1, 2017

jreback commented Oct 1, 2017

jreback commented Oct 1, 2017

jbrockmendel commented Oct 2, 2017

jreback commented Oct 3, 2017

jbrockmendel commented Oct 3, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback left a comment

Choose a reason for hiding this comment

jbrockmendel commented Oct 6, 2017

jreback commented Oct 6, 2017

jreback commented Oct 6, 2017

jreback commented Oct 6, 2017

jbrockmendel commented Oct 6, 2017

jreback commented Oct 6, 2017

codecov bot commented Sep 7, 2017 •

edited

Loading

codecov bot commented Sep 7, 2017 •

edited

Loading

pep8speaks commented Sep 24, 2017 •

edited

Loading